Premium
Deprescribing Statins—Is It Ethical?
Author(s) -
Strandberg Timo E.
Publication year - 2016
Publication title -
journal of the american geriatrics society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.992
H-Index - 232
eISSN - 1532-5415
pISSN - 0002-8614
DOI - 10.1111/jgs.14270
Subject(s) - medicine , statin , diabetes mellitus , adverse effect , confounding , disease , population , intensive care medicine , dementia , physical therapy , endocrinology , environmental health
dementia. Although Cintosun and colleagues note that a large number of participants and the ability to follow them for up to 4 decades strengthened our observational study, they present some methodological concerns, which we would like to address. We do not claim to make any causal inferences about random measured blood glucose levels and dementia-related death but point to a clear risk association. Cintosun and colleagues do not mention that we adjusted for education (related to socioeconomic status) and cardiovascular disorders but were unable to include data on hospitalizations. Adjusting for relevant confounders is important, but we do not view it likely that factors such as vitamin deficiencies or thyroid disorders would alter the association between glucose level and dementia-related death, as Cintosun and colleagues suggest. We performed a wide assessment with variables such as depressive symptoms, anxiety, and alcohol, including a sensitivity analysis restricting the follow-up to the age of 65 (early-onset dementia), and the results did not differ to a significant degree. Although the above-mentioned factors are deemed independent risk factors of dementia, their importance when evaluating high glucose levels and dementia-related death is uncertain. In response to Cintosun’s concern regarding the use of dementia-related death, we included all registered deaths with dementia on the death certificate as the underlying or accompanying cause, and in understanding the mechanism of cause-specific dementia mortality, the use of dementiarelated death as an outcome in observational studies is considered valid. Furthermore, a validation study of Norwegian death certificates in dementia in nursing home residents showed high specificity. Using data from a large case–control study of dementia-related death (n = 561), we found that the risk of dementiarelated death in individuals with various apolipoprotein E genotypes were comparable with risk ratios from studies using clinical diagnoses as the end point, suggesting that dementia-related death could prove useful as a proxy for clinical dementia diagnoses in epidemiological studies.